Hello all,
When I use point extractor (specifically I use [AKAZE](http://docs.opencv.org/3.0-beta/doc/tutorials/features2d/akaze_matching/akaze_matching.html) but it is similar to SIFT) I use the [FileStorage](http://docs.opencv.org/modules/core/doc/xml_yaml_persistence.html?highlight=filenode) to save it in a .XML file. After I extract the [points](http://docs.opencv.org/modules/features2d/doc/common_interfaces_of_feature_detectors.html?highlight=keypoint#keypoint), I manually give each point a class_id. Later on I save it in a xml file. After it is saved, I look at the xml file and everything is saved correctly. Then I try to load it and everything loads correctly except the class_id. I manually checked everything and only the class_id is wrong. Not only it is wrong but far exceeds the limits of a int (which the class_id is). For example
- I saved class_id = 1 and when I load
it I get class_id = 1065353216
- I saved class_id = 2 and when I load
it I get class_id = 1073741824
- I saved class_id = 3 and when I load
it I get class_id = 1077936128
Here is the code of how I load it
vector temp_kpts;
FileStorage fs("Myfile.xml", FileStorage::READ);
fs["KeyPoints_1"] >> temp_kpts;
cout<<"KeyPoitn ID is --> "<< temp_kpts[0].class_id < "<< temp_kpts[1].class_id < "<< temp_kpts[2].class_id <
↧